SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images

TL;DR

SyreaNet is a novel underwater image enhancement framework that combines synthetic and real data guided by a revised physical model and domain adaptation, improving generalization across diverse underwater conditions.

Contribution

This work introduces the first physically guided framework integrating synthetic and real underwater images with domain adaptation for enhanced UIE performance.

Findings

Outperforms state-of-the-art UIE methods qualitatively and quantitatively.

Effectively bridges domain gaps between synthetic and real underwater images.

Demonstrates robustness across various underwater conditions.

Abstract

Underwater image enhancement (UIE) is vital for high-level vision-related underwater tasks. Although learning-based UIE methods have made remarkable achievements in recent years, it's still challenging for them to consistently deal with various underwater conditions, which could be caused by: 1) the use of the simplified atmospheric image formation model in UIE may result in severe errors; 2) the network trained solely with synthetic images might have difficulty in generalizing well to real underwater images. In this work, we, for the first time, propose a framework \textit{SyreaNet} for UIE that integrates both synthetic and real data under the guidance of the revised underwater image formation model and novel domain adaptation (DA) strategies. First, an underwater image synthesis module based on the revised model is proposed. Then, a physically guided disentangled network is designed to predict the clear images by combining both synthetic and real underwater images. The intra- and inter-domain gaps are abridged by fully exchanging the domain knowledge. Extensive experiments demonstrate the superiority of our framework over other state-of-the-art (SOTA) learning-based UIE methods qualitatively and quantitatively. The code and dataset are publicly available at https://github.com/RockWenJJ/SyreaNet.git.